Long non-coding RNA FEZF1-AS1 facilitates cell proliferation and migration in colorectal carcinoma

Long non-coding RNAs (lncRNA) have been shown to play important roles in the development and progression of cancer. Here, we discovered a novel long noncoding RNA (lncRNA) FEZF1 antisense RNA1 (FEZF1-AS1) is markedly upregulated in human primary colorectal carcinoma (CRC) and associated with CRC metastasis and poor prognosis. Moreover, the downregulation of FEZF1-AS1 expression significantly inhibited the CRC cells proliferation, migration and invasiveness, suppressed S-phase entry in vitro, and repressed tumor growth and metastasis in vivo. In contrast, overexpression of FEZF1-AS1 could promote the aggressive behaviors of CRC cells. We further discovered that the downregulation of FEZF1-AS1 reduced its sense-cognate gene FEZF1 mRNA and protein expression in CRC cells. There was a positive correlation between FEZF1-AS1 and FEZF1 expression in CRC. Moreover, FEZF1 knockdown also significantly suppressed CRC cell proliferation, migration, and invasion. Our findings indicate that the dysregulation of FEZF1-AS1 participates in colorectal tumorigenesis and progression, which might be achieved, at least in part, through FEZF1 induction.


INTRODUCTION
Colorectal carcinoma (CRC) is one of the most common cancers in the world. In China, CRC ranks fifth among cancer deaths and its incidence is increasing continuously [1]. Though alterations in oncogenes and tumour suppressor genes have been reported in CRC [2], the precise molecular mechanisms underlying CRC pathogenesis remain to be fully elucidated. The long noncoding RNAs (lncRNAs) are a class of transcribed RNA molecules over 200 nucleotides with no proteincoding capacity [3]. Through regulating gene expression, lncRNAs have been shown to play crucial roles in multiple biological processes such as development, differentiation and carcinogenesis [4][5][6][7]. Dysregulated lncRNA expression has been reported in various types of cancers, such as bladder, prostate, lung, breast, gastric and colorectal cancers [8][9][10][11][12][13][14][15], indicating that lncRNAs may be involved in tumorigenesis or tumor progression. FEZ family zinc finger 1 antisense RNA 1 (FEZF1-AS1) which located on the opposite strand of gene FEZF1, is a recently identified long non-coding RNA. However, its function in normal or cancer cells remain unknowns. In this study, we first identified that the up-regulation of lncRNA FEZF1-AS1 was associated with aggressive phenotypes of CRC and the poor prognosis in patients with CRC. Further function experiments revealed that FEZF1-AS1 increase cell proliferation, migration and invasiveness. Moreover, FEZF1-AS1 knockdown reduced its sense-cognate gene FEZF1 mRNA and protein expression in CRC cells, and there was a positive correlation between FEZF1-AS1 and FEZF1 expression in CRC. Furthermore, we validated that FEZF1 knockdown also significantly suppressed CRC cell proliferation, migration, and invasion. Taken together, these results suggest that FEZF1-AS1 participates as a non-coding oncogene in CRC carcinogenesis and metastasis.

LncRNA FEZF1-AS1 is up-regulated in human CRC tissues and associated with metastasis
Using real-time RT-PCR, we detected the expression levels of FEZF1-AS1 in 34 pairs of CRC and adjacent non-neoplastic mucosa tissues. The results indicated that FEZF1-AS1 expression levels in tumor tissues of CRC patients were significantly higher than those in corresponding normal tissues (p = 0.004, Figure 1A). It was observed that the up-regulation of FEZF1-AS1 in tumor samples was associated with lymph-node metastasis to a significant extent (p = 0.041, Figure 1B). In patients diagnosed with lymph node metastases, the relative mean expression of FEZF1-AS1 was over 2.30 fold higher than that in patients without metastases (0.00076 ± 0.00020 vs.0.00033 ± 0.00006). This association indicated that FEZF1-AS1 might have a pivotal role in CRC metastasis.

Up-regulation of FEZF1-AS1 is associated with aggressive phenotypes of patients with CRC
To explore whether FEZF1-AS1 expression levels are associated with the clinicopathological factors of CRC, we measured FEZF1-AS1 expression in a large cohort of 153 archived paraffin-embedded CRC and normal colon tissues using in situ hybridization. We found FEZF1-AS1-specific staining was observed in the cytoplasm of benign and malignant epithelial cells ( Figure 1C). We also observed that 89/153 (58.17%) CRCs had highlevel expression of FEZF1-AS1, whereas 22/56 (39.29%) normal mucosa tissues had high-level expression of FEZF1-AS1 (p = 0.015). The correlation analysis between clinicopathological characteristics and FEZF1-AS1 level indicated that high-level expression of FEZF1-AS1 was significantly associated with T-stage (p = 0.005), lymph node metastasis (p = 0.004), and distant metastasis (p = 0.020) in patients with CRC. However, it was not associated with other clinical pathological features, including patients' age, sex, tumor site, tumor size, and tumor differentiation degree (p > 0.05, Table 1).

Up-regulation of FEZF1-AS1 is correlated with poor prognosis of patients with CRC
To further evaluate the prognostic value of FEZF1-AS1 in CRCs, we analyzed the association between FEZF1-AS1 expression and survival duration using Kaplan-Meier analysis with the log-rank test. The results revealed that high-level expression of FEZF1-AS1 in CRC was significantly correlated with overall survival (Log Rank = 9.333, p = 0.002, Figure 1D) and disease-free survival (Log Rank = 9.329, p = 0.002, Figure 1E) of CRC patients. The high-level of FEZF1-AS1 was associated with short survival time (53.53 ± 3.86 vs.79.40 ± 4.02).
To determine whether the expression of FEZF1-AS1 was an independent prognostic factor for CRC, univariate and multivariate analyses were performed. The results indicated that high-level expression of FEZF1-AS1 was considered as an independent prognostic factor of outcomes in patients with CRC (p = 0.035, Table 2).

FEZF1-AS1 promotes CRC cell proliferation and colony formation in vitro
We measured the expression level of FEZF1-AS1 in a panel of CRC cell lines (Figure 2A), and found that there were higher expression level of FEZF1-AS1 in HCT116, M5 and LOVO cell line. So, we firstly inhibited the endogenous expression of FEZF1-AS1 in M5 and HCT116 cells by shRNA to further investigate the biological effects of FEZF1-AS1 on CRCs. Moreover, FEZF1-AS1 was overexpressed in SW480 and DLD-1 cells using transfection of pcDNA-FEZF1-AS1 (p < 0.001, Figure 2B and 2C). The growth curves determined by CCK-8 assays indicated that cell proliferation was dramatically reduced by knockdown of endogenous FEZF1-AS1 expression in M5 and HCT116 cells, whereas overexpression of FEZF1-AS1 enhanced the proliferative capacity of SW480 and DLD-1 cells (p < 0.001, Figure 2D and 2E).
Further colony formation assays also showed that down-regulation FEZF1-AS1 expression could significantly inhibit the colony formation M5 and HCT116 cells. In contrast, overexpression of FEZF1-AS1 had the opposite effect (p < 0.05, Figure 2F and 2G). Thus, these data revealed that FEZF1-AS1 enhances CRC cell growth.

Knockdown of FEZF1-AS1 promotes G1 arrest, but does not cause apoptosis in CRC cells
To probe potential mechanisms by which FEZF1-AS1 enhanced CRC cell proliferation, we assessed cell cycle and apoptosis in CRC cell lines after FEZF1-AS1 knockdown or overexpression. Flow cytometric cell cycle assays in M5 and HCT116 cells demonstrated that FEZF1-AS1 knockdown led to a significant accumulation of cells at G0/G1-phase and a significant decrease in cells in S-phase (p < 0.05, Figure 2H). Conversely, the overexpressed FEZF1-AS1 mainly led to a reduction in the G0/G1 population and an increase in the S-phase (p < 0.05; Figure 2I). However, the proportion of apoptotic cells remained similar (data not shown). Thus, FEZF1-AS1induced promotion of CRC cells proliferation appeared to be mediated by modulation of the G1-S checkpoint, rather than by apoptosis.

FEZF1-AS1 promotes CRC cells migration and invasion in vitro
We also examined the effect of FEZF1-AS1 knockdown on CRC cells migration/invasion. Our Oncotarget 11273 www.impactjournals.com/oncotarget results showed that the FEZF1-AS1 knockdown inhibited cell invasion by 93.0% in M5 cells, and by 88.6% in HCT116, respectively (p < 0.001, Figure 3A) using Matrigel invasion assay. The wound-healing assay also illustrated that FEZF1-AS1 knockdown in CRC cells caused a significant decrease in cell migration (p < 0.001, Figure 3B). In contrast, invasion and migration were increased by FEZF1-AS1 overexpression in DLD-1 and SW480 cells (p < 0.001, Figure 3C and 3D). These results indicated that FEZF1-AS1 facilitated CRC cell migration and invasion in vitro.

Knockdown of FEZF1-AS1 inhibits tumor growth and metastasis in nude mice
The effect of FEZF1-AS1 on tumor growth was assessed in vivo by administering a subcutaneous injection of HCT116 with stable knockdown of FEZF1-AS1 and mock cells into the subcutaneous bilateral hind leg of mice. During a 30-day follow up period, the average tumor volume in HCT116 cells with FEZF1-AS1 knockdown was reduced compared with mock cells. Moreover, using immunohistochemical staining of resected tumor tissues for Ki-67, we found that the proliferation index was reduced in tumor tissues formed from FEZF1-AS1 knockdown, compared with those formed from control cells (p < 0.05, Figure 4A-4C).
The effect on metastasis was evaluated 8 weeks after the mice had been injected with the same cells through tail vein. We examined the number and size of tumor metastatic nodules under a microscope in the lung. As shown in Figure 4D and 4E, the number of pulmonary metastatic nodules was clearly decreased in the knockdown of FEZF1-AS1 group compared with that of in control group (p = 0.036,). These results indicate that FEZF1-AS1 downregulation could inhibit tumor growth and metastasis in vivo. Konckdown of FEZF1-AS1 decreases the expression of FEZF1 mRNA and protein FEZF1-AS1 is a conserved ~2.6-kb RNA transcribed from the plus strand of chromosome 7, on the opposite strand of the gene coding FEZF1 protein (7q31.32), including 611 nucleotides of full complementarity between the first FEZF1-AS1 exon and FEZF1 exons 1 ( Figure 5A). To determine the relationship between FEZF1-AS1 and FEZF1, we examined the expression levels between FEZF1-AS1 and its sense-cognate gene FEZF1. After knockdown endogenous FEZF1-AS1 expression in M5 and HCT116 cells, the expression of FEZF1 mRNA was reduced in both two cell lines. This was confirmed in protein level by western Oncotarget 11275 www.impactjournals.com/oncotarget blotting ( Figure 5B and 5D). However, down-regulated of FEZF1 expression did not significantly affect FEZF1-AS1 expression in CRC cells ( Figure 5C). Furthermore, we also detected the expression levels of FEZF1 in the same cohort of 34 paired tissue samples using real-time PCR, the results indicated that FEZF1 expression levels in tumor tissues of CRC patients were significantly higher than those in corresponding normal tissues (p < 0.05, Figure 5E). At the same time, we observed a good positive correlation between FEZF1-AS1 and FEZF1 mRNA expression (Spearman's correlation analysis, r 2 = 0.1346; p = 0.033, Figure 5F). These results suggested that the expression of FEZF1 might be regulated by FEZF1-AS1.

siRNA-mediated knockdown of FEZF1 inhibited CRC cell proliferation, migration and invasion in vitro
We then tested whether FEZF1 was functionally involved in CRC tumorigenesis. siRNA transfectionmediated FEZF1 knockdown decreased cell growth relative to negative control at day 5 in both cell lines (p < 0.05; Figure 6A and 6B). Furthermore, we also examined the effect of FEZF1 knockdown on M5 cells migration/ invasion using the wound-healing and matrigel invasion assay. The results indicated that FEZF1 knockdown in M5 cells caused a significant decrease in cell migration (p = 0.009 and 0.007, Figure 6C). Matrigel invasion assay also showed that the FEZF1 knockdown inhibited cell invasion by 59.8% and 54.9% in M5 cells (p = 0.002 and 0.003, Figure 6D). These results indicated that the downregulation of FEZF1 expression was sufficient in reducing both cell migration and invasion in vitro.

DISCUSSION
It has now become widely accepted that mammalian genomes encode numerous lncRNAs [16]. Dysregulation of some lncRNAs has been shown in various types of cancers, including CRC [5,8,11,[17][18][19][20][21][22]. However, the functions and the mechanisms behind lncRNAs in CRC remain obscure. We compared the different expression profiles of lncRNA between CRC and normal mucosa tissues and mined numerous lncRNAs with deregulation expression in CRC tissues (data not shown). Among those different expression lncRNAs, we chose and focused on lncRNA FEZF1-AS1, a previously unstudied lncRNA. To the best of our knowledge, we present the first study that evaluates the prognostic impact of FEZF1-AS1 expression and comprehensive function studies of FEZF1-AS1 in CRC.
In this study, we found that the expression of FEZF1-AS1 increased in CRC tissues, and CRC patients with higher FEZF1-AS1 expression tend to have advanced T-stage, lymph node, or distant metastasis. Furthermore, a Kaplan-Meier analysis revealed that FEZF1-AS1 over-expression in tumor cells had a significantly worse prognostic impact on the overall survival of CRC patients. The results indicated that FEZF1-AS1 could serve as a predictor for advanced CRC and poor prognosis for CRC patients.
In order to highlight the function of up-regulation of FEZF1-AS1 in CRC, we further explored the critical role of FEZF1-AS1 in the progression of CRC by lossof-function and gain-of-function experiments. Our results revealed that the reducing of FEZF1-AS1 expression could significantly inhibit CRC cells proliferation, migration and invasiveness, and suppress S-phase entry in vitro. In contrast, overexpression of FEZF1-AS1 could promote the aggressive behaviors of CRC cells. Furthermore, the downregulation of FEZF1-AS1 could inhibit the tumorigenesis and lung metastasis in murine model. Taken together, these findings suggested that FEZF1-AS1 might serve as an oncogene and play a promotion role in CRC development and progression. Many other lncRNAs have also been implicated in the development of multiple tumors and identified as cancer biomarkers. Examples include promotion of oesophageal adenocarcinoma cells invasion and metastasis by HNF1A-AS1, control of colorectal cancer cells apoptosis by ZFAS1,    reprogramming of induced pluripotent stem cells by RNA-RoR, and regulation of non-small cell lung cancer cells growth and apoptosis by MEG3 [22][23][24][25]. Thus, effective blocking of these lncRNAs in cancer could be a novel preventive and therapeutic strategy.
The importance of lncRNAs in human cancer may be associated with their abilities to impact cellular functions through various mechanisms. Previous studies have demonstrated that the mechanisms of lncRNAs are partly dependant on their genomic location [26,27]. It is noteworthy that some lncRNAs, which are oriented in antisense direction with respect to a protein coding loci in the opposite strand, usually act as regulator of this gene [28,29]. FEZF1-AS1 is localized in the antisense DNA strand of the FEZF1 gene, which has been identified as critical roles in gastric cancer cell proliferation and tumorigenesis by transcriptional activation of the K-ras gene [30,31]. Therefore, we hypothesized that FEZF1-AS1 might regulate FEZF1 to contribute to progression of malignant diseases. Further experiments revealed the knockdown of FEZF1-AS1 reduced FEZF1 mRNA and protein expression in CRC cells, and that the expression levels of FEZF1 were significantly higher in tumor tissues. Consistent to these finding, there was a positive correlation between FEZF1-AS1 and FEZF1 sense expression in CRC. These data showed that FEZF1 mRNA expression is under the control of lncRNA FEZF1-AS1. Furthermore, we validated that FEZF1 knockdown also significantly suppressed CRC cell proliferation, invasion, and migration, but did not affect the expression of FEZF1-AS1. Our observations indicated that the promotion effects of FEZF1-AS1 on CRC might be achieved partly by activating FEZF1 expression. Because of including 611 nucleotides of full complementarity between the exon of FEZF1-AS1 and FEZF1, we used an RNase protection assay to test the possibility of RNA duplex formation to analysis the mechanism that FEZF1-AS1 regulated FEZF1 expression. However, the results did not validated that FEZF1-AS1 formed RNA duplex with FEZF1 mRNA and increased stability of FEZF1 mRNA (data not shown). So, further studies are needed to elucidate the precise mechanisms by which FEZF1-AS1 modulates its targets.
In summary, this study provides the first link between FEZF1-AS1 expression and CRC development. We proved that the FEZF1-AS1 was upregulated in CRC tissues and its increased expression might play a promotion role in CRC tumorigenesis and progression.
These results indicate that FEZF1-AS1 might be a candidate prognostic biomarker and a target for new therapies of CRC.

Cell culture
The human embryonic kidney cells 293T and the human CRC cell lines DLD-1, HT29, HCT116, SW480, SW620, Lovo were obtained from a cell bank at the Chinese Academy of Sciences (Shanghai, China). In previous studies, we described a subclone named M5 with enhanced metastatic abilities in liver [32,33]. All CRC cell lines were cultured in RPMI 1640 medium (Gibco, Gaithersburg, MD, USA) with 10% fetal bovine serum (HyClone, Logan, USA) and 100 U/ml penicillin/ streptomycin (Gibco). They were maintained in a humidified chamber containing 5% CO 2 at 37°C. 293T was maintained in Dulbecco's modified Eagle's medium (DMEM) that was supplemented with 10% FBS.

Tissue preparation
Fresh and formalin-fixed, paraffin-embedded, colorectal tumor tissue samples were obtained from patients who were diagnosed with primary CRC. Elective surgery was carried out on these patients at Nanfang Hospital, Southern Medical University (Guangzhou, China). In total, 34 cases of fresh CRC tissue were freshly frozen in liquid nitrogen and stored at -80°C until further use. 153 cases of archived, formalin-fixed, paraffinembedded CRC tissue samples were collected and used in clinicopathological and prognostic investigation of FEZF1-AS1. A comprehensive set of clinicopathological data were recorded, including age, gender, size of primary tumor, tumor differentiation, T stage, lymph node metastasis, and distant metastasis. The stage of disease was determined according to the tumor size, lymph node, and metastasis (pTNM) classification system [34]. Complete follow-up, ranging from 1-96 months, was available for the cohort of 153 patients, and the median survival was 56 months. The use of tissues for this study has been approved by the ethics committee of Nanfang Hospital, Southern Medical University. Before using these clinical materials for research purposes, all the patients signed the informed consent. None of these patients received any preoperative chemotherapy or radiotherapy.

In situ hybridization (ISH) and evaluation of staining of FEZF1-AS1
FEZF1-AS1 expression was examined using in situ hybridization (ISH) in CRC and non-CRC paraffinembedded sections. Briefly, after dewaxing and rehydration, the samples were digested with proteinase K, fixed in 4% paraformaldehyde, hybridized with the 5′digoxinlabeled probe 5′-TTCGACTGTTTCCTTGACACTAC-3′ (aligns to nucleotides 2550-2572 of the FEZF1-AS1 Reference Sequence NR_036484) at 55°C overnight, and subsequently incubated for 30 minutes at 4°C with HRP. Diaminobenzidine (DAB) was used for performing color reactions. An unrelated sequence that does not match any known human gene was used as negative control.
The positive expression of FEZF1-AS1 was primarily detected in the cytoplasm. To assess the clinical characteristics of patients, the ISH stained tissue sections were reviewed and scored separately by two blinded pathologists. Based on both the intensity and proportion of FEZF1-AS1-positive cells, the staining scores were determined using a relatively simple, reproducible scoring method [33,35]. On a scale of 0 to 3, the staining intensity was scored as follows: 0 (no staining), 1 (weak), 2 (medium), or 3 (strong). The extent of staining is defined as the percentage of positive staining areas of tumor cells or normal colonic epithelial cells in relation to the whole tumor area or entire section for normal samples. The extent of staining was scored on a scale of 0 to 4 as follows: 0, 0%; 1, 1-25%; 2, 26-50%; 3, 51-75%; and 4, 76-100%. The sum of the staining-intensity and staining-extent scores was used as the final staining score for FEZF1-AS1 (0-7). For statistical analysis, a final staining score of ≥ 3 was considered to be high.

Transfection of colon cancer cells
DLD-1, SW480, HCT116 and M5 cells were seeded into six-well plates. DLD-1 and SW480 cells were transfected with the pcDNA-FEZF-AS1 or empty vector, two FEZF1 siRNAs and control siRNA then were transfected into the HCT116 and M5 cells using Lipofectamine (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions, Cells were harvested after 48 hours for qRT-PCR and western blot analyses.

Colony formation assay
The cells were plated in 6-well plates at 2 × 10 2 per well and maintained in RPMI1640 containing 10% FBS for 2 weeks. After 2 weeks, the cells were washed twice with PBS, fixed with methanol and stained with 0.5% crystal violet. The number of colonies whose diameter was greater than 150 μm was counted under a microscope [36].

Cell migration assay
Cell motility was measured with wound healing assay by measuring the movement of cells in a scraped, acellular area created by a 200 μL pipette tube, and the spread of wound closure was observed after 0 and 48 hours, respectively. Photographs were taken to assess the level of migration in each group of transfected cells. The migration was quantified by counting the total number of cells that migrated toward the original wound field.

Cell invasion assay
For invasion assay, matrigel-coated chambers (BD Biosciences, San José, CA, USA) containing 8 µm pores were used. Cells were seeded in the upper chambers (coated in matrigel) at 2 × 10 5 concentration in serum-free medium. The lower chamber of the transwell was filled with culture media containing 10% FBS as a chemoattractant. After these chambers were incubated at 37°C for 48 hours, non-invaded cells on the top of the transwell were scraped off with a cotton swab. The successfully translocated cells were fixed with 10% formalin. Then, they were stained with 0.1% crystal violet for 30 minutes and counted under a light microscope.

Flow cytometric analysis
Cells were seeded at a density of 1 × 10 6 cells/well in six well plates. After 24 h, cells were washed with PBS and fixed in ice-cold 70% ethanol for 1 h and then treated with 100 μL of 50 mg/L propidium iodide for 30 min at 4°C in the dark. The cell cycle profiles were assayed using the Elite ESP flow cytometer at 488 nm, and data were analyzed with the CELL Quest software (BD Biosciences, San Jose, CA, USA).

Animal experiments
Balb/C-nu/nu athymic nude mice (4-6 weeks old) were obtained from the Laboratory Animal Centre of Southern Medical University. For tumor growth assay, a total of 2 × 10 6 cells suspended of HCT116 with stable knockdown of FEZF1-AS1, or mock cells, were respectively injected into left and right bilateral hind leg subcutaneous of mice. The tumor size was measured using digital calipers every three days. After 24 days of monitoring, mice were sacrificed by cervical dislocation and tumors were dissected. Tumor volume was calculated www.impactjournals.com/oncotarget as follows: Volume = (D × d 2) /2, where D meant the longest diameter and d meant the shortest diameter.
For the in vivo metastasis assay, the mice were i.v. injected via the lateral tail vein with 2 × 10 6 tumor cells or control cells were suspended in 0.1 mL of culture medium. Animal health status was monitored every three days. Human endpoints had been planned at the end of experiment as a means to relieve pain or distress. At the time of euthanasia, the lungs were removed by dissection away from adjacent organs and fixed with 10% neutral buffered formalin. Subsequently, the consecutive tissue sections were obtained and stained with haematoxylineosin (H & E) to observe the metastatic nodules of lungs under microscope. All the experimental procedures were performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Committee on the Ethics of Animal Experiments of Southern Medical University. All the necessary steps were taken to minimize the suffering and distress caused to the mice.

Statistical analysis
All statistical analyses were performed using the SPSS 16.0 statistical software package. In at least three independent experiments, the data were presented in terms of mean ± SEM. The differences between variables were assessed using three statistical tests: χ 2 test, Fisher's exact test, or One-way ANOVA. For patients with different levels of FEZF1-AS1 expression, the survival curves were plotted using the Kaplan-Meier method and compared using the log-rank test. The multivariate survival analysis was performed on all parameters that were found to be significant in univariate analysis using the Cox regression model. A p value less than 0.05 was considered as statistical significance.

GRANT SUPPORT
This study was supported by National Nature Science Fund of China (No. 81272763, 81000953, 81172242 and 81472318) and the Natural Science Foundation of Guangdong Province (2014A030313286 and 2014A030313308).